Hospira (UK) Ltd v

For an obviousness challenge to a dosing regimen, the question is whether the skilled team would undertake the necessary work with a fair expectation of success, not whether it would know in advance that the regimen would work. A pharmacokinetic assessment may involve judgment and experience as well as mathematical modelling. Evidence that a higher dose would maintain a therapeutic concentration may make a lower claimed dose obvious where the evidence supports sufficient confidence in the therapeutic window. The court need not identify the precise lower boundary of an obvious range if the claimed dose falls within it on the balance of probabilities. The burden of proving obviousness remains on the challenger.

Genentech appealed against Birss J’s decision that its patent for an intravenous trastuzumab dosing regimen was invalid for obviousness, alternatively insufficiency and loss of priority. The patent claimed an initial dose of 8 mg/kg followed by at least two doses of 6 mg/kg at three-week intervals, differing from the FDA-approved weekly regimen.

The judge found that the skilled oncology team would consider three-weekly dosing, and that pharmacokinetic evidence based on a safely administered 500 mg dose supported a three-weekly regimen. Genentech argued that the evidence did not establish a sufficient basis for doses below 7.14 mg/kg because the judge had rejected part of the further modelling. The central issue was whether the accepted evidence made the claimed 8 + 6 q3w regimen obvious.

1. Appeal dismissed. The finding that the claimed regimen was obvious was upheld. The issues of insufficiency and priority therefore did not arise.
2. Under section 3 of the Patents Act 1977, the relevant question was whether the skilled team would embark on the necessary work with a fair expectation of success. The team did not need to know that the regimen would work. It needed sufficient confidence in its prospects to warrant a small, reasonably defined clinical trial.
3. The evidence accepted at trial showed that three-weekly dosing would occur to the skilled clinician. The pharmacokineticist could use the FDA label, including the safely administered 500 mg dose and its longer half-life, to conclude that a three-weekly regimen would remain above the target trough concentration. That conclusion was not merely arithmetic. Clinical pharmacokinetics involved judgment, experience, uncertainty and the practical therapeutic window.
4. The Court of Appeal held that the first set of calculations and the associated evidence were sufficient, without reliance on the rejected part of the further modelling, to establish that an obvious range included 8 + 6 q3w. The judge did not need to determine the precise lower limit of that range. It was enough that, on the balance of probabilities, the claimed dose fell within it.
5. The absence of evidence of a surprising advantage in choosing 6 mg/kg rather than 7.14 mg/kg was relevant to the assessment, but did not reverse the burden of proof. Hospira retained and discharged the burden of showing obviousness.
6. The court also accepted that a skilled team conducting a trial would be free to adjust doses within the therapeutic window while monitoring pharmacokinetic parameters. The appeal was dismissed, and the alternative grounds did not arise.

1. Court of Appeal (Civil Division) — [2015] EWCA Civ 57: appeal by Genentech dismissed.
2. High Court of Justice, Chancery Division — Birss J, [2014] EWHC 1094 (Pat): patent held invalid for obviousness, alternatively insufficiency and loss of priority; permission to appeal was refused.
3. Permission to appeal — subsequently granted on the papers by Lewison LJ.